alihan/llama-cpp-python
1
0
Fork 0
mirror of https://github.com/abetlen/llama-cpp-python.git synced 2023-09-07 17:34:22 +03:00
Code Issues Packages Projects Releases Wiki Activity

Set kv cache size to f16 by default

Browse Source
This commit is contained in:
Andrei Betlen
2023-04-14 22:21:19 -04:00
parent 9c8c2c37dc
commit 6e298d8fca
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
llama_cpp/llama.py
View File

@@ -21,7 +21,7 @@ class Llama:
n_ctx: int = 512, n_ctx: int = 512,
n_parts: int = -1, n_parts: int = -1,
seed: int = 1337, seed: int = 1337,
f16_kv: bool = False, f16_kv: bool = True,
logits_all: bool = False, logits_all: bool = False,
vocab_only: bool = False, vocab_only: bool = False,
use_mmap: bool = True, use_mmap: bool = True,